Components of Investment Performance (Fama, 1972)

Components of Investment Performance
Author(s): Eugene F. Fama

Source: The Journal of Finance , Jun., 1972, Vol. 27, No. 3 (Jun., 1972), pp. 551-567
Published by: Wiley for the American Finance Association
Stable URL: https://www.jstor.org/stable/2978261
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
and Wiley are collaborating with JSTOR to digitize, preserve and extend access to The Journal
of Finance
This content downloaded from

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
All use subject to https://about.jstor.org/terms
the Journal of FINANCE
VOL. XXVII JUNE 1972 No. 3
COMPONENTS OF INVESTMENT PERFORMANCE*
EUGENE F. FAMA4
I. INTRODUCTION
THIS PAPER SUGGESTS methods for evaluating investment performance. The

topic is not new. Important work has been done by Sharpe [21, 22], Treynor
[23], and Jensen [13, 14]. This past work has been concerned with measuring
performance in two dimensions, return and risk. That is, how do the returns
on the portfolios examined compare with the returns on other "naively
selected" portfolios with similar levels of risk?
This paper suggests somewhat finer breakdowns of performance. For ex-
ample, methods are presented for distinguishing the part of an observed return
that is due to ability to pick the best securities of a given level of risk
("selectivity") from the part that is due to predictions of general market price
movements ("timing"). The paper also suggests methods for measuring the
effects of foregone diversification when an investment manager decides to
concentrate his holdings in what he thinks are a few "winners."
Finally, most of the available work concentrates on single period evaluation
schemes. Since almost all of the relevant theoretical material can be presented
in this context, much of the analysis here is likewise concerned with the one-
period case. Eventually, however, a multiperiod model that allows evaluations
both on a period-by-period and on a cumulative basis is presented.
II. FOUNDATIONS
The basic notion underlying the methods of performance evaluation to

be presented here is that the returns on managed portfolios can be judged
relative to those of "naively selected" portfolios with similar levels of risk.
For purposes of exposition, the definitions of a "naively selected" portfolio
and of "risk" are obtained from the two-parameter market equilibrium model
of Sharpe [20], Lintner [15, 16], Mossin [18] and Fama [10, 11]. But it is
-well to note that the two-parameter model just provides a convenient and some-
what familiar set of naively selected or "benchmark" portfolios against which
* Research on this paper was supported by a grant from the National Science Foundation.
t Graduate School of Business, University of Chicago.
551

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
552 The Journal of Finance
the investment performance of managed portfolios can be evaluated. As indi-

cated later, other risk-return models could be used to obtain benchmark port-
folios consistent with the same general methods of performance evaluation.
In the simplest one-period version of the two-parameter model, the capital
market is assumed to be perfect-that is, there are no transactions costs or
taxes, and all available information is freely available to everybody-and
investors are assumed to be risk averse expected utility maximizers who believe
that return distributions for all portfolios are normal. Risk aversion and
normally distributed portfolio returns imply that the expected utility maxi-
mizing portfolio for any given investor is mean-standard deviation efficient.'
In addition, investors are assumed to have the same views about distributions
of one-period returns on all portfolios (an assumption usually called "homoge-
neous expectations"), and there is assumed to be a riskless asset f, with both
borrowing and lending available to all investors at a riskless rate of interest Rf.
It is then possible to show that in a market equilibrium all efficient port-
folios are just combinations of the riskless asset f and one portfolio of risky
assets m, where m, called the "market portfolio," contains every asset in the
market, each weighted by the ratio of its total market value to the total market
value of all assets. That is, if Rm, E(Rm) and O(Rm.) are the one-period return,
expected return, and standard deviation of return for the market portfolio
m, and if x is the proportion of investment funds put into the riskless asset f,
then all efficient portfolios are formed according to2
Rx =xRf + (1 - x)Rm x 1, (1)
so that
E(Rx) = xRf + (1 - x)E(Rm) (2)
c(Rx) = (1 - x)a(Rm). (3)
Geometrically, the situation is somewhat as shown in Figure 1. The curve

b m d represents the boundary of the set of portfolios that only include risky
assets. But efficient portfolios are along the line from Rf through m. Points
below m (that is, x > 0) involve lending some funds at the riskless rate Rf
and putting the remainder in m, while points above m (that is, x < 0) involve
borrowing at the riskless rate with both the borrowed funds and the initial
investment funds put into m.
In this model the equilibrium relationship between expected return and risk
for any security j is
E (R'j) Rf + r] cov(RRm) (Ex ante market line). (4)

a (R.) a(Rm)
Here coy v is the covariance between the return on asset j and the
1. By definition, a mean-standard deviation efficient portfolio must have the followin

No portfolio with the .same or higher expected one-period return has lower standard deviation of
return.
2. Tildes ('Y are used throughout to denote random variables. When we refer to realized values
of these variables, the tildes are dropped.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
Components of Investment Performance 553
E(R)
E (RM) - --- - - - A /-
R~~~~~~~~~~~~
f ~~b
in~~~~~~~~~~~~~i
L a (R)~(Rm
FIGURE 1
The Efficient Set with Riskless Borrowing and Lending
on the market portfolio m. In the two-parameter model a(R.) is a measure of

the total risk in the return on the market portfolio m. Since the only risky
assets held by an investor are "shares" of m, it would seem that, from a port-
folio viewpoint, the risk of an asset should be measured by its contribution to
(Y(Rm). In fact this contribution is just cov(Rj,Rm)/a(Ri,n). Specifically, if
xjm is the proportion of asset j, j 1,.., N, in the market portfolio m
O(Rrn) ZE iXjlncov(Rj~,
(Rm)
f,)(5)
In this light (4) is a relationship between expected return and risk which says
that the expected return on asset j is the riskless rate of interest Rf plus a risk
premium that is [E(Rnm) - Rf]/a(Rm), called the market price- per unit of
risk, times the risk of asset j, cov(Rj,m)/a(R"m.).
Equation (4) provides the relationship between expected return and risk
for portfolios as well as for individual assets. That is, if xj, is the proportion
N
of asset j in the portfolio p (so that xjp -1), then multiplying both sides of
jwg
(4) by x,p and summing over j, we get

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
FE(Rmn) - Rf cov(RP, Rm)

E(Rp) Rf + O(R) I O(Rm) (6)
where, of course,
N
_p EzxjpRj.
J=1
But (4) and (6) are expected return-risk relations derived under the as-
sumption that investors all have free access to available information and all
have the same views of distributions of returns on all portfolios. In short, the
market setting envisaged is a rather extreme version of the "efficient markets"
model in which prices at any time "fully reflect" available information. (See,
for example [7].) But in the real world a portfolio manager may feel that he
has access to special information or he may disagree with the evaluations of
available information that are implicit in market prices. In this case the
"homogeneous expectations" model underlying (4) provides "benchmarks"
for judging the manager's ability to make better evaluations than the market.
The benchmark or naively selected portfolios are just the combinations of
the riskless asset f and the market portfolio m obtained with different values
of x in (1). Given the ex post or realized return Rm for the market portfolio,
for the naively selected portfolios, ex post return is just
R,, xRr + (1-x) R, (7)

that is, (1) without the tildes. Moreover,3
cov (R,, Rni) cov ( [I - x] R,,, R'%m)

(3X - =(1 I- x) a(R,,,)=o(R.,). (8)
a ( R,,,,1 ) a ( Rn)-(R*(8
That is, for the benchmark portfolios risk and standard deviation of return
are equal. And the result is quite intuitive: In the homogeneous expectations
model these portfolios comnprise the efficient set, and for efficient portfolio
risk and return dispersion are equivalent.
For the naively selected portfolios, (7) and (8) imply the following relation-
ship between risk 1, and ex post return Rx:
Rx= Rr ( Rm. -) ) (ex post market line). (9)
That is, for the naively selected portfolios there is a linear relationship between
risk and return that is of precisely the same form as (4) except that the ex-
pected returns that appear in (4) are replaced by realized returns in (9).
In the performance evaluation models to be presented, (9) provides the
benchmarks against which the returns on "managed" portfolios are judged.
These "benchmarks" are used in a sequence of successively inore complex sug-
gested performance evaluation settings. First we are concerned with one-period
models in which a portfolio is chosen by an investor at the beginning of the
3. Henceforth the risk cov(Rj,R,n)/a(Rm) of an asset or portfolio j will be denoted as [j.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
period, its performance is evaluated at the end of the per

intermediate cash flows or portfolio decisions. Then we consider multiperiod
evaluation models that also allow for fund flows and portfolio decisions be-
tween evaluation dates. We find, though, that almost all of the important
theoretical concepts in performance evaluation can be treated in a one-period
context.
III. THE BENCHMARK PORTFOLIOS: SOME EMPIRICAL ISSUES
Before introducing the evaluation models, however, it is well to discuss some

of the empirical issues concerning the so-called "market lines" (4) and (9).
Since this paper is primarily theoretical, and since empirical problems are best
solved in the context of actual applications, the discussion of empirical issues
will be brief.
First of all, to use (9) as a benchmark for evaluating ex post portfolio re-
turns requires estimates of the risk, P and dispersion, a(Rp), of the managed
portfolios as well as an estimate of a (R1), the dispersion of the return on the
market portfolio. If performance evaluation is to be objective, it must be
possible to obtain reliable estimates of these parameters from historical data.
Fortunately, Blume's evidence [3, 4, 5] suggests that at least for portfolios of
ten or more securities, Phi and ca(Rp) seem to be fairly stationary over lon
periods of time (e.g., ten years), and likewise for o(R1km).
But other empirical evidence is less supportive. Thus throughout the analysis
here normal return distributions are assumed, though the data of Fama [6],
Blume [3], Roll [19] and others suggest that actual return distributions con-
form more closely to non-normal two-parameter stable distributions. It would
conceptually be a simple matter to allow for such distributions in the evalua-
tion models (cf. Fama [11]). But since the goal here is just to suggest some
new approaches to performance evaluation, for simplicity attention will be
restricted to the normal model.
Finally, the available empirical evidence (e.g., Friend and Blume [12],
Miller and Scholes [17], and Black, Jensen and Scholes [2]) indicates that
the average returns over time on securities and portfolios deviate systematic
from the predictions of (4). Though the observed average return-risk rela-
tionships seem to be linear, the tradeoff of risk for return (the price of risk)
is in general less than would be predicted from (4) or (9). In short, the evi-
dence suggests that (4) and (9) do not provide the best benchmarks for the
average return-risk tradeoffs available in the market from naively selected
portfolios.
Even these results do little damage to the performance evaluation models.
They indicate that other benchmark portfolios than those that lead to (9)
might be more appropriate, but given such alternative "naively selected"
portfolios, the analysis could proceed in exactly the manner to be suggested.
For example, Black, Jensen and Scholes [2] compute the risks (13's) for each
security on the New York Stock Exchange, rank these, and then form ten
portfolios, the first comprising the .iN securities with the highest risks and
the last comprising the .iN securities with the lowest risks, where N is the

ffff:ffff:ffff:ffff:ffff:ffff on Thu, 01 Jan 1976 12:34:56 UTC
total number of securities. They find that over various subperiods from
1931-65 the average monthly returns among these portfolios are highly cor-
related, and when plotted against risk the average returns on these portfolios
lie along a straight line with slope somewhat less than would be implied by the
"price of risk" in (4) or (9). As benchmarks for performance evaluation
models, their empirical risk-return lines seem to be natural alternatives to (9).
And with these alternative benchmarks, performance evaluation could proceed
precisely as suggested here. But again, for simplicity, we continue on with the
more familiar benchmarks given by (9).
It would be misleading, however, to leave the impression that all important
empirical problems relevant in the application of performance evaluation
models have been solved. To a large extent the practical value of such models
depends on the empirical validity of the model of market equilibrium-that is,
the expected return-risk relationship-from which the benchmark or "naively
selected" portfolios are derived. And though much interesting work is in
progress, it would be rash to claim that all empirical issues concerning models
of market equilibrium have been settled.
For example, an important (and unsolved) empirical issue in models of
market equilibrium is the time interval or "market horizon period" over which
the hypothetical expected return-risk relationship is presumed to hold. Does
the model hold continuously (instant by instant), or is the market horizon
period some discrete time interval? This is an important issue from the view-
point of performance evaluation since if the market horizon period is discrete,
evaluation periods should be chosen to coincide with horizon periods.
The evidence of Friend and Blume [12] and that of Black, Jensen, and
Scholes [2] suggests that meaningful relationships between average returns
and risk can be obtained from monthly data, while the evidence of Miller and
Scholes [17] indicates that this is not true for annual periods. Within these
broad bounds, however, the sensitivity of risk-return relations to the time
interval chosen remains an open issue.
But unsolved empirical questions are hardly a cause for disheartenment. It
is reasonable to expect that some of the empirical issues will be solved in the
process of applying the theory. And in any case, application of a theory in-
variably involves some empirical approximations. The available evidence on
performance evaluation, especially Jensen's [13, 14], suggests that the re-
quired approximations need not prevent even more complicated evaluation
models from yielding useful results.
IV. PERFORMANCE EVALUATION IN A ONE-PERIOD MODEL

WHEN THERE ARE No INTRAPERIOD FUND FLOWS
Let Vat and Vat+i be the total market values at t and t + 1 of the actual
(a = actual) portfolio chosen by an investment manager at t. With all port-
folio activity occurring at t and t + 1, that is, assuming that there are no
intraperiod fund flows, the one-period percentage return on the portfolio is
Ra -Vat+ - Vat
Vat

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
One benchmark against which the return Ra on the chosen

compared is provided by Rx(Pfa), which by definition is
combination of the riskless asset f and the market portfolio
equal to (ia, the risk of the chosen portfolio a. One measure o
of the chosen portfolio a is then
Selectivity = Ra - Rx(Pa). (10)
That is, Selectivity measures how well the cho

naively selected portfolio with the same level of r
Selectivity, or some slight variant thereof, is th
in the work of Sharpe [21, 22], Treynor [23] an
detailed breakdowns of performance are possible
Overall
Performance Selectivity Risk
[RaRf] = [Ra-Rx(f(a)1 + [Rx((a)-Rf]. (11)
That is, the Overall Performance of the portfolio decision is the difference
between the return on the chosen portfolio and the return on the riskless
asset. The Overall Performance is in turn split into two parts, Selectivity (as
above) and Risk. The latter measures the return from the decision to take on
positive amounts of risk.4 It will be determined by the level of risk chosen
(the value of Pa) and, from (9), by the difference between the return on the
market portfolio, Rm, and the return on the riskless asset, Rf.
These performance measures are illustrated in Figure 2. The curly bracket
along the vertical axis shows Overall Performance which in this case is positive.
The breakdown of performance given by (11) can be found along the vertical
line from p,,. In this example, Selectivity is positive: A portfolio was chosen
that produced a higher return than the corresponding "naively selected"
portfolio along the market line with the same level of risk. Risk is also positive,
as it is whenever a positive amount of risk is taken and the return on the
market portfolio turns out to be higher than the riskless rate.
A. Selectivity: A Closer Look

If the portfolio chosen represents the investor's total assets, in the mean-
variance model the risk of the portfolio to him is measured by a (Rak),
standard deviation of its return. And the risk of the portfolio to the investo
a(Ra), will be greater than what might now be called its "market risk," f,a, a
long as the portfolio's return is less than perfectly correlated with the retur
on the market portfolio. To see this, note that the correlation coefficient ka
between Ra and Rm is
4. For greater descriptive accuracy, we should, of course, say "return from risk" or even "ret
from bearing risk," rather than just Risk. Likewise, "return from selectivity," would be more
descriptive than Selectivity. But (hopefully) the shorter names save space without much loss of
clarity.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
cov(Ra, ur)
(Y (REt) (y (Rm)
It follows that
cov(Ra, Rm)
Pa 11-1 = kamll (Ra)
so that 13a <Y(Ra) depending on whether kaen 1 -.5
R =5 + ( R m)f f
x f 1.' , x
x a _
R __ Net Selectivity
a
Selectivity D j - Selectivity
0 p Oi3aa
?T t (R ) ~~~~~~~~~~ - f, o(R)
FIGURE 2
An Illustration of the Performance Measures of Equations (11), (12), and (13).
Intulitively to some extent the portfolio decision may have involved puttin
more eggs into one or a few baskets than would be desirable to attain portfol
efficiency-that is, the manager places his bets on a few securities that he
thinks are winners. In other words, to the extent that G(ka) > pa, the portfolio
manager decided to take on some portfolio dispersion that could have been
diversified away because he thought he had some securities in which it would
pay to concentrate resources. The results of such a decision can be evaluated
in terms of the following breakdown of Selectivity:
Selectivity Diversification
[Ra Rx(PR) ]3 Net Selectivity + fRxt(G a)) - Rx(a)f; (12a)

or
Selectivrity Diversification
Net Selectivity = [R.-Rx(Pe)] -fRX(O(Re)) Rx(13a)J. (12b)
By definition, Rx(acRa)) iS the return on the combination of the

asset f and the market portfolio m that has return dispersion equ
5. In fact the naively selected portfolios are the only ones whose returns are literally perfectly
correlated with those of the market portfolio (cf. equation (8)). But the theoretical work of Fama
[91 and the empirical work of Black, Jensen and Scholes [2] suggests that the return on any
well-diversified portfolio will be very highly correlated with Rnm.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
that of the actual portfolio chosen. Thus Diversification measures the extra
portfolio return that the manager's winners have to produce in order to make
concentration of resources in them worthwhile. If Net Selectivity is not posi-
tive, the manager has taken on diversifiable risk that his winners have not
compensated for in terms of extra return.
Note that, as defined in (12), Diversification is always non-negative, so that
Net Selectivity is equal to or less than Selectivity. When R11 > R5, Diversifica-
tion measures the additional return that would just compensate the investor
for the diversifiable dispersion (that is, a(Ra) - Pa) taken on by the manager.
When R,1 < Rf (so that the market line is downward sloping), Diversification
measures the lost return from taking on diversifiable dispersion rather than
choosing the naively selected portfolio with market risk and standard deviation
both equal to Pa, the market risk of the portfolio actually chosen.
The performance measures of (12) are illustrated in Figure 2 along the
dashed vertical line from (Ra). In the example shown, Selectivity is positive
but Net Selectivity is negative. Though the manager chose a portfolio that
outperformed the naively selected portfolio with the same level of market risk,
his Selectivity was not sufficient to make up for the avoidable risk taken, so
that Net Selectivity was negative.
The breakdown of Selectivity given by (12) is the only one that is con-
sidered here. The rest of Section IV is concerned with successively closer
examinations of the other ingredient of Overall Performance, Risk. Before
moving on, though, we should note that (12) itself is only relevant when di-
versification is a goal of the investor. And this is the case only when the
portfolio being evaluated constitutes the investor's entire holdings, and the
investor is risk averse. For example, an investor might allocate his funds to
many managers, encouraging each only to try to pick winners, with the investor
himself carrying out whatever diversification he desires on personal account.
In this case Selectivity is the relevant measure of the managers' performance,
and the breakdown of Selectivity of (12) is of no concern.
B. Risk: A Closer Look

If the investor has a target risk level PT for his portfolio, the part of
Overall Performance due to Risk can be allocated to the investor and to the
portfolio manager as follows:
Risk Manager's Risk Investor's Risk
A A A -5
[Rx(Pa)- RfI [Rx(pa) RX(PT)] + [RX(PT) - Rf] (13)

RX(PT) is the return on the naively selected portfolio with the target l
market risk. Thus Manager's Risk is that part of Overall Performance a
Risk that is due to the manager's decision to take on a level of risk Pa d
from the investor's target level PT, while Investor's Risk is that part o
Performance that results from the fact that the investor's target level
is positive. These performance measures are illustrated in Figure 2 a
dashed vertical line from PT.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
Manager's Risk might in part result from

at least the manager might have chosen a po
or lower than the target level because he fel
do abnormally well or abnormally poorly du
ation. But if an estimate of E(Rm.) is availab
results of such a timing decision can be obt
the ex ante market line of (4)7 we can subd
Risk Manager's Timing
[Rx(pa) - Rf] = {[Rx(pa) - E(Rx(Pa))] [RX(PT)- E(RX(T))]}

Total Timing Market Conditions
+ [E(Rx(pa)) -E(Rx(PT))] + [Rx(PT)

Manager's Exp. Risk Investor's Risk
The first three terms here sum to the Manager's Risk of (13). M
Expected Risk is the incremental expected return from the manager
to take on a nontarget level of risk. Market Conditions is the difference be-
tween the return on the naively selected portfolio with the target level of risk
and the expected return of this portfolio. It answers the question: By how
much did the market deviate from expectations at the target level of risk?
Total Timing is the difference between the ex post return on the naively
selected portfolio with risk (3n and the ex ante expected return. It is positive
when Rm > E(Rm) (and then more positive the larger the value of Ps), and it
is negative when R.1 < E(R1.) (and then more negative the larger the value
of PIn). The difference between Total Timing and Market Conditions is Man-
ager's Timing: it measures the excess of Total Timing over timing performance
that could have been generated by choosing the naively selected portfolio with
the target level of risk. Manager's Timing is only positive when the sign of
the difference between Pa and P( is the same as the sign of the difference
between R1m. and E(Rn1), that is, when the chosen level of market risk is above
6. E(,11) might be estimated from past average returns on the market portfolio m. Alternatively,
past data might be used to estimate the average difference between Rm and Rf. In any case, it
should become clear that the expected values used must be naive or mechanical estimates (or at least
somehow external to those being evaluated), otherwise the value of the timing measures is
destroyed.
Admittedly, given the current status of empirical work on the behavior through time of averag
returns on risky assets, we can at most sepeculate about the best way to estimate E(Rm). Hopefully
empirical work now in progress will give more meaningful guidelines. And perhaps the developmen
of theoretical methods of performance evaluation will itself stimulate better empirical work on
estimation procedures. In any case, the discussion in the text should help to emphasize that one
cannot obtain precise measures of returns from timing decisions without mechanical or naive
estimates of equilibrium expected returns.
7. That is,
[E(Rm) f
E(R (sm))ilaRr for (( Da
and similarly for E (;Rx(PT))

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
(below) the target level and Rl.Jis above (below) E (nR,). It

more sensitive than Total Timing as a measure of the results of a timing
decision.
A target level of risk will not always be relevant in evaluating a manager's
performance. For example, an investor may allocate his funds to many man-
agers, with the intention that each concentrates on selectivity and/or timing,
with the investor using borrowing or lending on personal account to attain his
desired level of market risk.
If a target level of risk is not relevant but the expected value or ex ante
market line is still available, a breakdown of Risk similar to (14) can be ob-
tained by treating the market portfolio (or the appropriate proxy)8 as the
target portfolio. That is,
Risk Manager's Timing

A - - A_ - _ I a
[Rx(pa) Rf] {
Total Timing Market Conditions
+ [E(Rx(Pa)) -E(Rn)] + [Rm Rf]. (15)
Expected Deviation Market Risk

from Market
The idea here is that even in the absence of a target level of risk, the measure
of Manager's Timing must be standardized for the deviation of the market
return from the expected market return, that is, for the "average" spread
between the ex post and ex ante market lines.
Finally, the goal of this paper is mainly to suggest some ways in which
available theoretical and empirical results on portfolio and asset pricing models
can provide the basis of useful procedures for performance evaluation. But
the various breakdowns of performance suggested above are hardly unique..
Indeed any breakdown chosen should be tailored to the situation at hand. For
example, if a target level of risk is relevant but the subdivision of Risk given
by (14) is regarded as too complicated, then the approximate effects of the
timing decision might still be separated out as follows:
Risk Total Timing

f % f-A A
[Rx(Pa) - Rf Rx(PR) E(Rx((a) ) ]

Manager's Expected Risk Investor's Expected Risk
+ [E(Rx(Pa)) -E(Ax(3T))] + [E(Rx(PT)) - RfJ. (16)

The one new term here is Investor's Expected Risk, which measures the ex-
pected contribution to Overall Performance of the investor's decision to have
a positive target level of risk. Alternatively if a target level of risk is not
8. For example, if one were faced with portfolio evaluation in a multiperiod context, one might
use the average of past levels of market risk chosen by the manager as a proxy for the target risk
level when the latter is not explicitly available.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
relevant for the situation at hand, but an expected value line is available, Risk
can nevertheless be subdivided as follows,
Risk Total.Timing Total Expected Risk
A A A
[Rx(Pa)- Rf] - [Rx((3a) - E(Rx(Pa))] + [E(Rx(Pa)) - Rf]. (17)

And these few suggestions hardly exhaust the possibilities.
V. COMPONENTS OF PERFORMANCE: MULTIPERIOD MODELS

WITH INTRAPERIOD FUND FLOWS
In the one-period evaluation model presented above, (i) the time at which
performance is evaluated is assumed to correspond to the portfolio hori
date, that is, the time when portfolio funds are withdrawn for consumption;
and (ii) there are assumed to be no portfolio transactions or inflows and out-
flows of funds between the initial investment and withdrawal dates, so that
there is no reinvestment problem. If in a multiperiod context we are likewise
willing to assume that: (i) though there are many of them, evaluation date
nevertheless correspond to the dates when some funds are withdrawn for con-
sumption, and (ii) all reinvestment decisions and other portfolio transactions
are also made at these same points in time, then generalization of the one-
period model to the multiperiod case is straightforward.9 Indeed the basic
procedure could be period-by-period application of the performance measures
presented in the one-period model. The major embellishments would not be in
the nature of new theory, but rather would arise from the fact that multiperio
performance histories allow statistically more reliable estimates of the various
one-period performance measures.
But this pure case is unlikely to be met in any real world application. Often
performance evaluation would be carried out by someone with little or no
knowledge of the dates when funds are needed for consumption by the owner
of the portfolio, and often (e.g., in the case of a mutual fund or a pension
fund) the portfolio is owned by many different investors with different con-
sumption dates. As a result evaluation dates, withdrawal dates, and reinvest-
ment dates do not usually coincide.
The rest of this paper is concerned with how the concepts of the one-period
model must be adjusted to deal with such intraevaluation period (or more
simply, intraperiod) fund flows. The procedure is to first present detailed
definitions of variables of interest in models involving intraperiod fund flows,
and then to talk about actual measures of performance. And it is well to keep
in mind that though the analysis is carried out in a multiperiod context, the
problems to be dealt with arise from intraperiod fund flows. With such fund
flows, the same problems would arise in a one-period evaluation model.
A. Definitions
Suppose the investment performance of a portfolio is to be evaluated at
discrete points in time, but that there can be cash flows between evaluation
9. For the development of the underlying models of consumer and market equilibrium for this
case see [8].

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
dates. That is, there can be intraperiod inflows in the form o

receipts (dividends, interest) on existing portfolio holdings or
tributions of capital by new or existing owners. And there can be
outflows in the form of dividend payments to the portfolio's owner(s) (e.g., a
mutual fund declares dividends) or withdrawals of capital (e.g., by a mutual
fund's shareholders).
In simplest terms, the major problem with intraperiod cash flows is obtain-
ing a measure of the return on the beginning of period market value of a
portfolio that abstracts from the effects of intraperiod new contributions and
withdrawals on the end of period value of the portfolio. One approach is what
might be called the mutual fund method. Specifically, when performance
evaluation is first contemplated, the market value of the portfolio is subdivided
into "shares." Subsequently, whenever there are contributions of new capita
or withdrawals of capital from the portfolio, the current market value of a
share is computed and the number of shares outstanding is adjusted to reflect
the effects of the cash flow.10
Thus let evaluation dates correspond to integer values of t and define
V'a,t = actual market value of the portfolio at time t. It thus includes the
effects of investment of new capital or reinvestment of any cash in-
come received on securities held in the portfolio, and it is net of
any dividends paid out to owners or other withdrawals of funds prior
to t.
Va,t market value the portfolio would have had at t if no dividends were
paid out to owners since the previous evaluation date. In computing
Va,t it is simply assumed that dividends paid to the portfolio's owners
were instead reinvested in the entire portfolio. At the beginning of
each evaluation period, however, Va,t is set equal to V',,t.
nt= number of shares outstanding in the portfolio at t. As indicated
above, this is adjusted when new capital comes into the portfolio
and when capital is withdrawn, but it is unaffected by reinvestment
of cash income received on securities held or by dividends paid to
the portfolio's owners.
P'a,t V'a,t/nt actual market value at t of a share in the portfolio.
Pa,t = Va,t/nt = value of a share at t under the assumption that dividends
paid to owners of the portfolio were instead reinvested in the entire
portfolio.
Ra,t= (Pa,t - Pa,t_-)/pa,t_-. Assuming t corresponds to an evaluation
date, this is the one-period return on a share with reinvestment of
all dividends paid on a share since the last evaluation date.
Ra,t is an unambiguous measure of the return from t - 1 to t on a dollar

invested in the portfolio at t - 1. This is not to say, however, that it is un-
affected by intraperiod fund flows. Such fund flows are usually associated with
redistributions of portfolio holdings across securities and these affect the
10. This is in fact the method of accounting used by open end mutual funds. It is also closely
related to the "time-weighted rate of return" approach developed by Professor Lawrence Fisher.
On this point see [1, Appendix I and p. 2181.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
return on a share. Moreover, Ra t as defined above is not the only unambiguous

measure of the return from t - 1 to t on funds invested in the portfolio at
t - 1. For example, one could define Rat = (P'a,t + dt - P'a,t-i)/P'a,t-1,
where dt is the dividend per share paid during the evaluation period to the
portfolio's owners. The more complicated definition, that is, with dividends
assumed to be reinvested, is '(purer" (especially for the purpose of inter-
portfolio comparisons of performance) in the sense that funds invested at the
beginning of a period remain invested for the entire period, but it is less pure
in the sense that it assumes a reinvestment policy not actually followed in the
portfolio.
The next step is to define prices per share for the benchmark or naively
selected portfolios that also take account of intraperiod fund flows.
Pxt(PT) = price at t per share of the naively selected portfolio with the
target risk level. To avoid double-counting of past performance,
at the beginning of any evaluation period (for example, just after
an evaluation takes place at t - 1) this price is set equal to the
price per share of the actual portfolio. Then this amount is in-
vested in the naively selected portfolio with the target risk level,
and the behavior of the market value of this portfolio during the
evaluation period determines the end-of-period price per share,
Pxt(PT). Any intraperiod cash income generated by the securities
of this naively selected portfolio is assumed to be reinvested in
this portfolio.
These conventions for the treatment of beginning-of-period values and int
period cash income will be taken to apply in the definitions of all the benchma
portfolios. Thus
pt(Rf) =price at t per share of the naively selected portfolio obtained by
investing all funds available at t - 1 in the riskless asset.
The benchmarks provided by Pxt(PT) and pt(Rf) are unaffected by intra-
period fund flows in the actual portfolio. This is not true of the following
two benchmarks.
Pxt(Pa) = price at t per share of the naively selected portfolio with

market risk equal to that of the actual portfolio. At the
beginning of any evaluation period and after any transaction
in the actual portfolio during an evaluation period (that is,
after any cash flow or exchange of shares in the actual port-
folio) the market risk of the actual portfolio is measured, and
the current price per share of this benchmark is shifted into
the naively selected portfolio with that level of market risk.
Thus the value of 15a could be shifting more or less contin-
uously through time as a result of inflows and outflows of
funds and decisions to shift the holdings in the portfolio.'
11. Indeed even if there are no transactions taking place, the value of Oa shifts continuo
through time as a result of shifts in the relative market values of individual securities in the
portfolio. Aside from adjusting the value of D1a at the beginning of each evaluation period, we hav
chosen to ignore the effects of such "non-discretionary" shifts here.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
Pxt(O(Ra)) =price at t per share of the naively selected

turn dispersion equal to that of the actual portfolio. The
definition of Pxt('(R,)) is obtained by substituting O(Ra
Pa, in the definition of P.;t(Pa) above.
Thus p.t(f3) and pt(o(A)) take account of changes in P.,, and o(

result from intraperiod fund flows and portfolio shifts. Computationally, keep-
ing tract of P(3n and o(Ra) in the way required for these benchmarks is not a
difficult problem. At any point in time the market risk (3n of the chosen port-
folio is just the weighted average of the market risks of the individual assets
in the portfolio, where the weights are the proportions of total portfolio market
value represented by each asset. Thus if one has estimates of the market risks
of the assets from which portfolios are chosen, the value of P,, is updated by
combining these with current measures of the weights of individual assets in
the chosen portfolio. And a similar procedure can be followed with respect to
updating values of a (Rp) 12
B. Multiperiod Measures of Performance

Given the beginning and end-of-period prices per share for these benchmark
portfolios, their one-period returns are obtained in the usual way. Then the
performance history of a portfolio can be built up (for example) through
period-by-period application of the breakdowns given by (11)- (13). Alterna-
tively, one can define performance measures in terms of profit per share rather
than return. Thus, in line with (13) and using end of evaluation period prices,
define
Overall
Performance Selectivity
[Pa,t Pt(Rf)] = [Pat- Pxt((a)]
Manager's Risk Investor's Risk

A - A
+ [Pxt(Pa) -Pxt(PT)] + [Pxt(PT) -pt(Rf)] (18)
This type of breakdown can of course be computed both period-by-period and

cumulatively. And from such multiperiod histories one can get more reliable
measures of a portfolio manager's true abilities than can be obtained from a
one-period analysis. For example, one can determine whether his Selectivity
is systematically positive or simply randomly positive in some periods.
For some purposes one may wish to compare the multiperiod performance
histories of different portfolios. For example, an investment company may
be interested in the relative abilities of its different security analysts and
portfolio managers. Or an investor who has allocated his funds to more than
one manager may be interested in comparing their performances. On a period-
by-period basis such performance comparisons can be carried out in terms of
percentage returns. Alternatively, if the prices of shares in different portfolios
12. Keeping track of o(Va) is especially simple if one assumes. that returns are generated by
the so-called "market model." On this, and for additional computational suggestions, see Blume
[3, 4, 51.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
are set equal at the beginning of comparis

measures such as (18) could be computed both on a period-by-period basis
and cumulatively.
One must not get the impression, however, that all the problems caused by
intraperiod fund flows have been solved. Though the performance of a "share"
during any given evaluation period (or across many periods) gives an un-
ambiguous picture of the investment history of funds invested in a given
portfolio at a given point in time, comparisons of the performances of shares
in different portfolios are not completely unambiguous. This is due to the fact
that even when things are done on a per share basis, intraperiod fund flows
necessitate portfolio decisions that usually have some effect on the performance
of a share. And when such fund flows occur at different times (and thus
during different market conditions) in different portfolios, the observed per-
formances of shares in the portfolios may differ, even if the portfolios are
managed by the same person trying to follow the same policies in all of his
portfolio decisions. But though such ambiguities seem unavoidable and to
some extent unsolvable, their effects on performance comparisons should be
minor except in cases where portfolios experience large cash flows (relative to
their total market values) in short periods of time and/or when evaluation
periods are long.
Finally, if an ex ante market line is available to compute expected values
through time for the three benchmarks, pxt(P5), Pxt(Pa) and p.t(O(Ra)), then
the one-period performance breakdowns of (14)-(17) can be carried out
either in terms of returns or market values, and these can be used as the basi
of even more detailed multiperiod performance histories.
But we terminate the discussion at this point. We do this not because of a
lack of additional interesting problems, but because in the absence of actual
applications, suggested solutions become increasingly speculative and thus of
less likely usefulness.
VI. SUMMARY
Some rather detailed methods for evaluating portfolio performance have

been suggested, and some of the more important problems that would arise in
implementing these methods have also been discussed. In general terms, we
have suggested that the return on a portfolio can be subdivided into two parts:
the return from security selection (Selectivity) and the return from bearing
risk (Risk). Various finer subdivisions of both Selectivity and Risk have also
been presented.
To a large extent the suggested models can be viewed as attempts to com-
bine concepts from modern theories of portfolio selection and capital market
equilibrium with more traditional concepts of what constitutes good portfolio
management.
For example, the return from Selectivity is defined as the difference be-
tween the return on the managed portfolio and the return on a naively selected
portfolio with the same level of market risk. Both the measure of risk and the
definition of a naively selected portfolio are obtained from modern capital
market theory, but the goal of the performance measure itself is just to test
how good the portfolio manager is at security analysis. That is, does he show

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC
any ability to uncover information about individual securities that is not

already implicit in their prices?
Likewise, traditional discussions of portfolio management distinguish be-
tween security analysis and market analysis, the latter being prediction of
general market price movements rather than just prediction of the special
factors in the returns on individual securities. The various timing measures
suggested in this paper provide estimates of the returns obtained from such
attempts to predict the market. And modern capital market theory again
plays a critical role in defining these estimates.
REFERENCES
1. Bank Administration Institute. Measuring the Investment Performance of Pe

Ridge, Illinois: B.A.I., 1968.
2. Fisher Black, Michael Jensen, and Myron Scholes. "The Capital Asset Pricing Model: Some
Empirical Tests." To appear in Studies in the Theory of Capital Markets, edited by Michael
Jensen and published by Praeger.
3. Marshall Blume. "The Assessment of Portfolio Performance," unpublished Ph.D. dissertation,
University of Chicago, 1968.
4. . "Portfolio Theory: A Step Toward Its Practical Application." Journal of Busi-
ness XLIII (April, 1970), 152-173.
5. . "On the Assessment of Risk," Journal of Finance XXVI (March, 1971), 1-10.
6. Eugene F. Fama. "The Behavior of Stock Market Prices," Journal of Business X:XXVIII
(January, 1965), 34-105.
7. . "Efficient Capital Markets: A Review of Theory and Empirical Work," Journal
of Finance XXV (May, 1970), 383-417.
8. . "Multiperiod Consumption-Investment Decisions," American Economic Review
XL (March, 1970), 163-174.
9. . "Portfolio Analysis in a Stable Paretian Market," Management Science XII
(January, 1965), 404-419.
10. - -. "Risk, Return and Equilibrium: Some Clarifying Comments," Journal of
Finance XXIII (March, 1968), 29-40.
11. . "Risk, Return, and Equilibrium," Journal of Political Economy LXXIX
(January-February, 1971), 30-55.
12. Irwin Friend, and Marshall Blume. "Measurement of Portfolio Performance under Uncertainty,"
American Economic Review XL (September, 1970), 561-575.
13. Michael Jensen. "The Performance of Mutual Funds in the Period 1945-64," Journal of
Finance XXIII (May, 1968), 389-416.
14. . "Risk, the Pricing of Capital Assets, and the Evaluation of Investment Port-
folios," Journal of Business XLII (April, 1969), 167-247.
15. John Lintner. "Security Prices, Risk, and Maximal Gains from Diversification," Journal of
Finance XX (December, 1965), 587-615.
16. . "The Valuation of Risk Assets and the Selection of Risky Investments in Stock
Portfolios and Capital Budgets," Review of Economics and Statistics XLVII (February,
1965), 13-37.
17. Merton Miller, and Myron Scholes. "Rates of Return in Relation to Risk: A Reexamination of
Some Recent Findings." To appear in Studies in the Theory of Capital Markets, edited by
Michael Jensen, and published by Praeger.
18. Jan Mossin. "Equilibrium in a Capital Asset Market," Econometrica XXXIV (October, 1966),
768-783.
19. Richard Roll. The Behavior of Interest Rates: Application of the Efficient Market Model to
U.S. Treasury Bills. New York: Basic Books, Inc., 1970.
20. William F. Sharpe. "Capital Assets Prices: A Theory of Market Equilibrium under Conditions
of Risks," Journal of Finance XIX (September, 1964), 425-442.
21. . "Mutual Fund Performance," Journal of Business XXXIX (Special Supplement,
January, 1966), 119-138.
22. -. "Risk Aversion in the Stock Market," Journal of Finance XX (September,
1965), 416-422.
23. Jack L. Treynor. "How to Rate Management of Investment Funds," Harvard Business Revie
XLIII (January-February, 1965), 63 75.

193.80.252.177 on Sun, 06 Mar 2022 12:19:09 UTC

Components of Investment Performance (Fama, 1972)

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Components of Investment Performance (Fama, 1972)

Uploaded by

Copyright:

Available Formats

Components of Investment Performance

Author(s): Eugene F. Fama

Stable URL: https://www.jstor.org/stable/2978261

This content downloaded from